Michael P
- Research Program Mentor

PhD candidate at University of California Berkeley (UC Berkeley)

Expertise

AI & machine learning, deep learning (neural networks), mathematical analysis, applications of AI.

Bio

Artificial intelligence today is driven by the "deep neural network", and while they've shown tremendous feats, many of their problems get swept under the rug: (a) It's incredibly hard to research/develop a neural network model for a new idea/application, and also often incredibly expensive. (b) Even if you get a neural network to work on your data, these models are often quite "brittle" and break down when put in the real world. My goal is to address both of these problems, through a mathematical understanding of these models and a practical eye to make them reliably work in the real world. Outside of research: my biggest hobby is piano. Played since I was 3, started purely classical and transitioned to jazz/improv. Competed in middle school and played some restaurant gigs in high school, but now I just play/compose for fun to keep my artistic side healthy. If I'm not playing piano, I'm most likely running, playing Smash/chess/insert-game-here, or snowboarding.

Project ideas

Project ideas are meant to help inspire student thinking about their own project. Students are in the driver seat of their research and are free to use any or none of the ideas shared by their mentors.

AI to write your running playlist

There are plenty of music-generating AI already available, but these are all trying to please everyone and make generically good-sounding music. The idea here is make a music-generating AI for you, to combine industry knowledge and your knowledge on the music you love to make AI that actually sounds good.

Coding skills

Python (numpy, pytorch, JAX), MATLAB, C/C#

Teaching experience

Co-head instructor for CS 70 at Berkeley (giving lectures for undergrad class of ~370 students), high school tutor, avid practitioner and spokesperson of the Feynman technique for learning.

Credentials

Work experience

Moovila Inc (2018 - 2018)

Software engineer

Education

Princeton University

BA Bachelor of Arts

Mathematics (analysis and differential geometry), minors in CS and Applied Math

University of California Berkeley (UC Berkeley)

MS Master of Science candidate

AI, robotics, control